How companies game the ETSI standard-essential patent database to manipulate perceptions of leadership in 5G

Last time, I explained that to understand leadership in the technology standard that defines 5G, it is necessary to look at the foundational importance of individual inventions and what they bring to the standard. Only by evaluating these two factors can one get a full, accurate picture of 5G and the different building blocks that have contributed to its development. Approaches that solely focus on simplistic metrics, such as counting the number of contributions made by individual participants, ignore the profound variance of inventions within the full technology tree of 5G. 

Despite this fact, many studies still use various quantitative measures as shortcuts to try to determine 5G leadership. One of the most common sources they draw from is the European Telecommunications Standards Institute (ETSI) 5G standard-essential patent (SEP) database. In this blog, I’ll take a deeper dive into exactly why this approach in particular is fundamentally flawed.

Standards are collaborative systems to develop useful technology solutions 

Technology standards like 5G are built through voluntary collaboration among industry-leading companies, with thousands of engineers from all over the world coming together to discuss current technical challenges, present the available solutions, and select the optimal technology choices based on technical merit.

In 5G, in addition to participating in this exchange of ideas, companies are asked in parallel to voluntarily declare what intellectual property rights (IPRs) they own, covering relevant technologies that are or may become essential to the standard. Declarations are made to the ETSI’s IPR database, which makes available a public repository of the disclosures.

Some level of over-declaration is natural and in good faith

Due to the nature of the standards process, it is understandable that some level of inaccuracy is unavoidable in these declarations. Until it gets finalized, the standard is a moving target — everything is being proposed, discussed and debated in real time, and the specifications iterate and evolve quickly. 

This means that ideas submitted, and the associated IP that was declared to be a potential part of the standard, may become less relevant as the standard grows and evolves in new directions over time. It is worth noting that ETSI does not provide a way to withdraw a declaration when the patent later becomes irrelevant to the 5G standard.

This results in some level of over-declaration as a “natural” part of the process, done in good faith with the aim of forward-looking transparency in mind.

The argument is simple: the more 5G SEPs a company has disclosed, the more important that company must be to the overall standard. Because of the spread of this and other quantitative approaches, companies are becoming incentivized to increase their disclosure numbers in the ETSI database, and some companies eager to boost their own leadership perceptions have recently taken to deliberately inflating their 5G IPR contributions in several ways.

Questionable tactics are being used to artificially inflate SEPs numbers

I want to highlight three questionable declaration behaviors that “game” these reports and distort the SEP landscape in the ETSI database:

Over-declaration of irrelevant IPRs

This is when a patent or patents clearly not relevant to 5G are disclosed as essential to the 5G standard. We’ve seen certain companies who were major innovators in 4G fall behind in 5G innovation, but simply declare almost all their old 4G patents as relevant to the new 5G standard, regardless of whether or not those technologies were a part of the new 5G specifications.

It is true that certain features born in 4G are fundamentally important to the way modern cellular communications work and are therefore also fundamental in 5G (and the associated IPRs thus apply to both 4G and 5G). A great example of this is the orthogonal frequency-division multiplexing (OFDM)-based waveform design that forms the basis of over-the-air transmission for both systems, as our own 5G leadership blog describes. However, some features in 4G do not carry over into 5G, and those IPRs are not relevant to the new 5G standard, so shouldn’t be declared as essential to 5G in the ETSI database.

Over-declaration at the family-level

Patents are property rights, but the bodies that grant the rights are inherently territorial (USPTO in US, EPO in Europe, CNIPA in China, etc.). Companies that do business globally need to patent the same invention in different jurisdictions to protect their inventions across countries. This results in what is called a “patent family” — the same or similar invention, patented separately in multiple different regions.

ETSI’s IPR Policy takes this into account and encourages only a single patent in a patent family to be declared as standard-essential. A company therefore typically chooses one patent from this family to declare to the standard, the so-called “basis patent.”

But some companies have recently taken the approach of separately declaring multiple patents from the same family (in effect the same invention) to the 5G standard. This results in increasing their counts of “basis” standard-essential patents, but it obviously doesn’t reflect an increased number of inventions that the company has contributed to the standard. However, some IP analysis firms continue to use the number of “basis” patents as a metric to determine 5G leadership.

Over-declaration at the 3GPP specification level

5G is made up of hundreds of different technical specifications, each defining how a particular aspect of the technology works. While the same invention can apply to a couple of specifications (for example, lower-layer aspects and upper-layer aspects), it is unlikely that the same invention applies to half a dozen technical specifications each focusing on a very narrow domain. Yet, some companies are now disclosing one patented invention as essential to several different specifications as a matter of common practice.

In the example below, taken from the publicly available ETSI forms, one can see a single patent declared as essential to 17 different specifications.  Looking closer at these specifications, they are from multiple working groups, which define entirely different parts of the system.

It should be noted that some IP analytics firms report SEP counts and charts for each individual specification. When that is the case, declaring the same SEP to many different specifications may result in that same SEP being counted two, three, four or even many more times in those counting exercises, which obviously paints a distorted picture as to the number of SEPs that a given company truly holds.

Increasing transparency and determining true leadership

These misleading declaration practices inflate the number of SEPs in the ETSI database and distort understanding of real leadership in 5G. This is an unfortunate side effect of the openness and transparency of the ETSI IPR process — a system based on good-faith self-declarations aimed at making it easier for all to see what IPRs may apply to the cellular standards, and one never intended as a metric to evaluate standard leadership.    

Unfortunately, IP analytics firms continue to base conclusions of 5G technology and IP leadership on this data. These fundamentally flawed conclusions are often picked up and quoted by popular media as well, and even used by policymakers in their analysis of the standards landscape.

Just as $100 bill is more valuable than a $1 bill, one cannot simply count how many pieces of paper you hold to understand true value. Shortcut counting methods are easy to understand but don’t provide an accurate method to determine leadership.

Both industry players and regulators have noticed the issues with using the ETSI database to determine leadership in 5G. For example, the recent proposed SEP regulation from the European Commission includes a portion about an alternate IP disclosure database with the goal to increase transparency. Qualcomm supports efforts to increase transparency and is actively working with industry partners both inside and outside of ETSI to achieve this goal.

In the meantime, as we have shown, counting declarations in the ETSI database, especially when they can be “gamed” in the above misleading ways, is not a reliable method to determine leadership. If one wants to understand true leadership in 5G, quality of inventions remains the single most important factor.